1. Forum
  2. Usenet
  3. LINUX.DEBIAN.BUGS.DIST

  • Bug#1109676: linux-image-6.12.35+deb13-amd64: Regression in host kernel

    From Salvatore Bonaccorso@21:1/5 to Jonathan on Wed Jul 23 15:50:02 2025
    XPost: linux.debian.kernel

    Hi Jonathan,

    On Mon, Jul 21, 2025 at 08:42:50PM +0100, Jonathan wrote:
    Package: src:linux
    Version: 6.12.35-1
    Severity: normal
    Tags: upstream
    X-Debbugs-Cc: [email protected], [email protected]
    User: [email protected]
    Usertags: amd64

    Dear Maintainer,

    I run desktop with iGPU and can start qemu VM(s) with a A750.
    It worked (reasonably) for few years and runs with 6.12.33 kernel.
    See start of VM boot (efi gop)
    Windows 11 guest now starts and power saves monitor. Device has problem code 43. Can ssh.
    A debian testing guest sticks to blank with cursor, from gdm, without being able to switch VT. Newest testing kernel fine inside with old host kernel. ssh
    in fine.

    Not fixed in 6.12.39 source.
    Bisected to; https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=linux-6.12.y&id=fb5873b779dd5858123c19bbd6959566771e2e83

    I also wonder if bugs #1109116 and #1109203 could be from same. Although I can
    still boot host onto card so not necessarily matching.

    thanks for pinpointing the introducing commit. Can you report the
    regression yourself to upstream and report back here the upstream
    report or do you need help with that?

    Include the regressions and stable list as well in the report, in
    addtion to the involved maintainers and persons of the breaking
    commit.

    Regards,
    Salvatore

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Salvatore Bonaccorso@21:1/5 to Jonathan Howard on Wed Jul 23 23:50:01 2025
    XPost: linux.debian.kernel

    Control: tags -1 - moreinfo
    Control: forwarded -1 https://lore.kernel.org/linux-iommu/[email protected]/T

    Hi,

    On Wed, Jul 23, 2025 at 08:00:57PM +0000, Jonathan Howard wrote:
    I spent time figuring out from it main patch was initially working.
    Just as I read the recommended upstream report method while composing an email; I looked on the maintainer list.

    Spotted related post. Same regression bisect.

    https://lore.kernel.org/linux-iommu/[email protected]/T

    Can confirm the patch correction fixes my regression.
    Just waiting for queue.

    Thank you! So it should land in 6.12.y upstream soonish. Right now we
    might defer a fix to the first trixie point release or the first DSA.

    Regards,
    Salvatore

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Salvatore Bonaccorso@21:1/5 to Salvatore Bonaccorso on Thu Jul 24 15:30:01 2025
    XPost: linux.debian.kernel

    Hi

    On Wed, Jul 23, 2025 at 11:38:45PM +0200, Salvatore Bonaccorso wrote:
    Control: tags -1 - moreinfo
    Control: forwarded -1 https://lore.kernel.org/linux-iommu/[email protected]/T

    Hi,

    On Wed, Jul 23, 2025 at 08:00:57PM +0000, Jonathan Howard wrote:
    I spent time figuring out from it main patch was initially working.
    Just as I read the recommended upstream report method while composing an email; I looked on the maintainer list.

    Spotted related post. Same regression bisect.

    https://lore.kernel.org/linux-iommu/[email protected]/T

    Can confirm the patch correction fixes my regression.
    Just waiting for queue.

    Thank you! So it should land in 6.12.y upstream soonish. Right now we
    might defer a fix to the first trixie point release or the first DSA.

    FTR, this fix has been included in the 6.12.40 release of today.

    Regards,
    Salvaotre

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)